Abstract:The prosperous development of the Internet of Things (IoT) and wireless communication technologies has led to explosive growth in the number of IoT sensor devices. However, some sensor devices are inevitably deployed in remote and inaccessible areas. How to continuously and reliably power sensor devices is a critical problem that needs to be addressed. Deploying self-powered modules on sensor devices by adopting self-powered technology is an effective solution to the energy shortage of sensor devices. Besides, Mobile Edge Computing (MEC) as a promising paradigm has provoked widespread popularity. With the help of MEC, devices can offload computing tasks to edge servers for processing, which greatly alleviates the limitations on energy, storage, and computation capability of devices. In this paper, we jointly study task scheduling and resource allocation in the MEC scenario where the sensor devices are with self-powered modules. Our goal is to minimize the long-term average energy consumption of self-powered sensor devices while ensuring system performance. We adopt stochastic optimization techniques to transform the modeled stochastic problem into a deterministic problem. Then, the deterministic problem is decomposed into four sub-problems, and we propose a task scheduling and resource allocation (TSRA) algorithm to solve these problems. Finally, we carry out a series of parameter analysis and comparison experiments to verify the TSRA algorithm. The experimental results show that our TSRA algorithm can make a dynamic tradeoff between energy consumption and system performance. It also demonstrates the effectiveness of our TSRA algorithm compared with other baseline algorithms.

Bayesian Prediction-Based Energy-Saving Algorithm for Embedded Intelligent Terminal

Terminal Security Monitoring Based on Power Consumption Information

An Energy-Saving Unbiased Gray-Prediction Algorithm of Internet of Things

An effective state-based predictive approach for leakage energy management on embedded systems

Adaptive energy saving algorithms for Internet of Things devices integrating end and edge strategies

Real-time Detection of Energy Consumption of IoT Network Nodes Based on Artificial Intelligence.

Optimal Task Scheduling and Resource Allocation for Self-Powered Sensors in Internet of Things: An Energy Efficient Approach

BER-based Power Scheduling in Wireless Sensor Networks

Genetic Algorithm Based Idle Length Prediction Scheme for Dynamic Power Management

An efficient algorithm for energy harvesting in IIoT based on machine learning and swarm intelligence

A Short-Term Energy Prediction System Based on Edge Computing for Smart City

A Novel Energy-Efficient Approach Based on Clustering Using Grey Prediction in WSNs for IoT Infrastructures

Data-driven Software-based Power Estimation for Embedded Devices

A Prediction-Based Charging Policy and Interference Mitigation Approach in the Wireless Powered Internet of Things.

Energy Loss Prediction in IoT Energy Services

Indoor–Outdoor Energy Management for Wearable IoT Devices With Conformal Prediction and Rollout

Utilizing Direct Photocurrent Computation and 2D Kernel Scheduling to Improve In-Sensor-Processing Efficiency

On-line remaining energy prediction: A case study in embedded battery management system ☆

Prediction-based Dynamic Energy Management in Wireless Sensor Networks

A Quantitative Study of Energy Consumption for Embedded Security.

Energy Consumption Optimisation Based on Mobile Edge Computing in Power Grid Internet of Things Nodes